. Opportunistic infections (OI) are the single most important cause of morbidity and mortality in people with AIDS. For this reason, prophylaxis and treatment of multiple opportunistic infections with a battery of anti-OI drugs is common clinical practice. However, this therapeutic strategy is limited by the high frequency of dose-limiting adverse reactions or therapeutic failures, many of which can be attributed to metabolic drug-drug interactions. Thus, there is an urgent need to predict, a priori, the drug combinations which are likely to interact in vivo so that this information can be used to either alter the combination of drugs used or to institute measures such as therapeutic drug monitoring, adjustment of doses, and close clinical monitoring of the patient. The overall aim of this proposal is to test the hypothesis that in vivo drug-drug interactions can be predicted from in vitro model systems such as human liver microsomes and hepatocytes. To do so, the investigators will determine if these in vitro model systems can predict (both qualitatively and quantitatively) in vivo inhibitory and inductive drug-drug interactions with respect to the metabolism of two clinically important anti-OI drugs - sulfadiazine (SDZ) to its toxic metabolite, N-hydroxySDZ, and clarithromycin (CLA) to its active metabolite, 14(R)-hydroxy clarithromycin. Specifically, the proposal will test the following hypotheses:1. Inhibitory drug-drug interactions: 1a.) SDZ and CLA are metabolized to their respective toxic (N-hydroxy) and active metabolites by cytochrome P450s (CYP) 2C9 and/or 3A4. 1b.) Anti-OI drugs (including fluconazole), which are potent inhibitors of CYP3A4 and 2C9, will significantly inhibit the in vitro metabolism of SDZ and CLA to their respective toxic and active metabolite. 1c.) The in vitro inhibition by fluconazole of N-hydroxylation of SDZ and 14(R) hydroxylation of CLA will qualitatively and quantitatively correlate with that obtained in vivo. 2. Inductive drug-drug interactions: 2a.) Rifabutin induces both CYP3A4 and 2C9 in human hepatocytes but it does so with a lower potency than rifampin. 2b.) Rifabutin is a less potent inducer than rifampin, both in vitro (in human hepatocytes) and in vivo, of N-hydroxylation of SDZ and, (R) hydroxylation of CLA. 2c.) The in vivo and in vitro potency of induction by rifabutin of N-hydroxylation of SDZ, sulfamethoxazole and dapsone will decrease in the following order: sulfamethoxazole- SDZ>dapsone.